--Renny Greenstone (rgreenst@pop900.gsfc.nasa.gov), Hughes STX Corp.
The Investigators Working Group (IWG) of the Mission to Planet Earth/Earth Observing System (MTPE/EOS) met for three days at the Mission Valley Marriott Hotel, San Diego, California--February 25-28, 1997.
Tuesday Morning, February 25
Ghassem Asrar, EOS Program Scientist, opened the meeting with a brief introduction. He said that many of last year's newly selected winners in the competition to be EOS investigators were with us at the meeting, and that this was the first opportunity for many of us to meet them and hear about their work. The focus of this first day's sessions would be on the early MTPE/EOS science, and the following morning (Wednesday) would be the occasion for reviews of programmatics. The afternoon would be set aside for poster presentations by the AM-1 and PM-1 instrument teams and also by the newly selected investigators. The final day's sessions (Thursday) would again be devoted to early MTPE/EOS science along with a special presentation updating alternative plans for the CHEM-1 mission.
Asrar then introduced Mike Mann, NASA's Deputy Associate Administrator for MTPE. Mann gave the good news that MTPE is doing much better than is reported by the press. The program funding has been stable throughout the past year. The issues confronting us are changing as we approach the first launch, TRMM, in November of this year. He said that it is good that we now have "real" results from MTPE of the sort to be reported at this meeting. Good scientific results justify support for the program-they keep the program alive.
Mike Freilich (Oregon State University), speaking for himself and Tim Liu (JPL), presented "NASA Scatterometer Measurements of Wind, Land, and Ice: Early Science Results." It was 14 years from the time the NASA Scatterometer (NSCAT) project was approved until the first results came down from the Japanese ADEOS satellite.
Freilich explained that NSCAT gives near-surface vector winds over oceans as its primary product. However, it also provides useful information, with better than 10-km resolution, through backscatter over land and ice. The major activity of the science team at this time is in the area of calibration/validation of the data, but he would not be discussing this aspect of the work in this talk. Rather he would devote most of his time to discussing the data products that have already been received and analyzed.
First, he reviewed the chronology of the ADEOS/NSCAT effort, starting with the ADEOS launch on August 17, 1996. The wind observation mode began on September 15; the first data were released to the science team on November 18; and the release of the data to the public occurred on February 24, 1997. Reprocessing of the data is to begin in March 1997, and a second reprocessing will begin in February 1998.
Freilich gave a quick review of the physical principles involved in NSCAT's wind determinations. NSCAT emits Ku band pulses at 14 Ghz. The pulses are scattered from cm-scale waves on the ocean surface with the scattering cross section, [[sigma]]0, an increasing function of wind speed. It is also angular dependent, with a maximum in the 0- and 180-degree directions. NSCAT has a 600-km swath coverage on either side of the orbit ground track, with a 300-km gap at nadir between the two swaths. Resolutions of 6-10 km have been achieved over land. A six-day image over Antarctica showed the difference between sea ice and glacier ice.
An interesting application of NSCAT over land was a study conducted during the eruption of a subsurface volcano over Iceland last September. In images on three-day centers the scattering cross section can be seen to increase, then drop, then rise again. All of this imaging was done under conditions of total cloud cover.
NSCAT can also be used to classify vegetation. Jungle areas are known to be isotropic homogeneous scatterers. Data collected for Amazonia were used to distinguish jungle from woodlands, from shrubs, and from grassland.
A comparison of Seasat data with NSCAT data for Amazonia (an 18-year separation between observations) brought out the development of a new reservoir and the presence of new settlement areas.
Freilich showed some examples of ocean wind data. Over the oceans, 25-to-50-km resolutions have been obtained. An image of NSCAT winds brought out the presence of both Typhoons Violet and Tom over the Pacific Ocean. Sequences of images brought out the transition of Typhoon Tom into an extratropical storm.
In a high-resolution mode, where the swath width is reduced to 300 km, but the resolution is increased to 12.5 km, Freilich showed the special advantages offered by NSCAT in an area where there is little conventional coverage. He showed data that had been obtained for South Georgia Island, in the South Atlantic to the east of Argentina, and compared his findings with the analysis provided by the National Centers for Environmental Prediction (NCEP). The NSCAT data showed the winds veering sharply to the left to run parallel to the length of the island (this was missed in the NCEP analysis). Freilich then performed a scale analysis, taking into account the presence of twelve peaks on the island, to show on theoretical grounds, that "upwind shadows" are to be expected with the sort of topography characterizing the island.
In weeks to come, Freilich expects to be providing air/sea-interaction study results. Looking further ahead, he said that SeaWinds on ADEOS II will be going up at the end of the century and will also have available water vapor information from the Japanese Advanced Microwave Scanning Radiometer (AMSR) instrument.
Freilich said that he could benefit from Tropical Atmospheric Ocean (TAO) buoy measurements to assist with NSCAT wind measurements in determining the 3-D wind fields, but that there is no sharp cut-off point as to where the supplementary data would be needed. He also pointed out that wind speed errors increase dramatically at wind speeds below 5 m/sec. (Tim Liu added that wind directions mean little at such low wind speeds.) Another point was that NSCAT wind speeds are significantly more accurate than Seasat winds.
Carl Wunsch (Massachusetts Institute of Technology) described "Science Accomplishments for TOPEX/Poseidon Mission." The mission was launched over four years ago with strong international participation. Areas of improvement to Earth science knowledge have included ocean tides, gravity field, orbit determination techniques, understanding of scattering from rough surfaces, and understanding of ionospheric structure.
Wunsch said that TOPEX/Poseidon is the first true global tide gauge. As a result of the mission, tidal elevations are now known to ~ 1 cm almost everywhere. As another consequence of the mission we now have rapid improvements in determinations of tidal dissipation rates. (Tidal dissipation dominates the evolution of the Earth/moon system, and is responsible for changing the moon's orbital characteristics.)
The TOPEX/Poseidon data allow the determination of global mean sea level variations. We can determine sea level changes at the scale of 2 mm/year. The data confirm a correlation between sea surface temperatures (SST) and sea level changes. The data also make clear that the ocean currents are not well represented by a static picture of the "conveyor belt," as has been described by Broecker.
TOPEX/Poseidon was able to achieve 1-to-2-cm accuracy in determining the shape of the ocean surface, in contrast to the ~10-m accuracy of previous determinations.
In a series of charts, Wunsch showed improvements in determinations of changes in temperature with depth in the oceans, leading to major corrections to model calculations of fresh-water fluxes and heat fluxes. The steps leading to the improvements involved adding in results from acoustic tomography of the oceans, GCMs, NCEP winds, and, finally, the TOPEX/Poseidon altimetry.
Within about a year, Wunsch said, it will be possible to have estimates of the 3-D time-evolving ocean circulation. It will be possible to calculate fluxes of biochemical constituents such as carbon, methane, and other nutrients.
He concluded his formal presentation by saying that with four years of TOPEX/Poseidon data it has been possible to carry oceanography from the geological era, in which the ocean currents are regarded as moving slabs, to something more like meteorology, in which daily patterns are viewed, and there is a predictive capability.
In answer to a question, he said that a currently available global synthesis of Earth's gravity field uses the best available geoid, and that the importance of the geoid varies on a case-to-case basis. Answering another question, he said that prior calculations using atmospheric residuals to determine oceanic heat fluxes between the equator and the poles are wrong. It may be necessary to achieve 5-km resolution over the oceans to get the right answers.
Charles Keeling(Scripps Institution of Oceanography) presented "Enhanced Plant Growth in the Northern High Latitudes," speaking for himself and Ranga Myneni (Boston University). He had been looking at signs of increased plant growth in the northern high latitudes, which may be related to global warming as the result of the carbon dioxide greenhouse effect. In particular, he has found evidence for advances in the time of the beginning of the growing season of plants in the high latitudes.
Keeling showed the monthly variation of the carbon dioxide cycle in the atmosphere for the period 1957-1995. The amplitude of the seasonal cycle has been increasing and has, in fact, increased by 17% over the past 20 years. He has used vegetation index data from the Pathfinder AVHRR analyses and also the GIMMS data supplied by Jim Tucker of the Goddard Space Flight Center.
One of the problems he has encountered, in looking for trends in the data, has been the apparent change in the calibration of the NOAA-7,-9, and -11 series spacecraft. Also, their equator crossing times have shifted over the years.
There was a comment from the audience that atmospheric interference over forests in northern New England has made vegetation index determinations from that region very difficult.
Pat McCormick (Hampton University) and P. K. Bhartia (Goddard Space Flight Center) gave a joint presentation on "Aerosol Measurements from Space: Current State-of-the-Art." McCormick led off with a short tutorial on sizes and characteristics of aerosols in the atmosphere, pointing out that they can be very regional, and that they can have stratospheric lifetimes of about one year.
Aerosol particles can cause changes in the number concentrations of cloud condensation nuclei (CCN) thus leading to smaller cloud particles as revealed, for instance, in the study of ship tracks evidenced in cloud images.
McCormick gave the lineage of the increasingly sophisticated passive spaceborne sensing of aerosols, which began with SAM II on Nimbus-7 (1 aerosol channel), followed by SAGE I on the AEM-2 satellite (4 aerosol channels), and then ending (so far) with SAGE II on the ERBS satellite (7 channels). He then went on to discuss the use of lidar for active sensing of aerosols, referring specifically to the Space Shuttle experiment known as LITE (Lidar in-space Technology Experiment). LITE had 30 nsec pulses and produced a 280-m footprint on the ground from the Shuttle.
It took ten years from starting point to implementation of the LITE Shuttle experiment. LITE used old-technology laser equipment and only functioned in the nadir direction (there was no scanning). Many interesting results were achieved with LITE in the short flight time of the Shuttle. By chance it was possible to detect the eye of Typhoon Melissa. Other observations clearly showed the presence of biomass burning over South America, and aerosol trajectories, from source on downstream, clearly brought out urban pollution plumes.
McCormick said that the recent IPCC report had singled out the low confidence now felt by the scientific community in the aerosol contribution to radiative forcing. He feels that improvements will come through combining lidar measurements with measurements from an oxygen A-band spectrometer (ABS) and also through using measurements from other instruments on board the EOS PM-1 platform. Later on, he feels that even more improvements will come from a complementary instrumented spacecraft that he called PICASSO. It would be ideal to have both passive and active measurements for tropospheric aerosol retrievals.
In the EOS era, SAGE III will be adding a lunar capability to the solar capability now offered by SAGE II for occultation measurements. SAGE III will have an 800-channel linear array.
P. K. Bhartia discussed new techniques for using the TOMS instrument to detect tropospheric aerosols. He said there have now been 18 years of TOMS data, starting in October 1978. In his new method, he uses the difference in absorption between 340 and 380 nm radiances. For the method to work, he must eliminate interfering cloud signals. He demonstrated the method, showing results of tracking the cloud from the Mt. St. Helen's eruption. He is now working to achieve quantitative estimates of the aerosol amounts. This requires determining the aerosol altitude, which he proposes to do by making use of the "Ring" effect. Ultimately, he believes that that there should be a new instrument dedicated to the operational measurements of aerosols and he is working on this. He has used lidar altimetry to confirm his altitude estimates.
V. Ramanathan (Scripps Institution of Oceanography) presented "Past Progress and Future Challenges," speaking for himself and Bruce Wielicki (Langley Research Center). Ramanathan gave a little historical perspective, saying that Samuel Pierpont Langley had invented the bolometer and that his measurements had been used by Svante Arrhenius to understand the Earth's radiation budget (ERB). Ramanathan referred to the difficulty posed by the angular-sampling bias in radiative flux determinations from satellites.
In his work, he has been trying to learn the effect of clouds on ERB. He stated that ERBE gave the first quantitative estimate of the net radiative effect of clouds on the radiation budget. ERBE showed that clouds on an annual and diurnal average basis led to a net radiative cooling of about 18 W m-2.
He asked what happens when Top-of-the-Atmosphere (TOA) measurements are combined with surface radiative energy measurements. In more-recent work using surface data collected over the western Pacific warm pool, Ramanathan (in addition to Robert Cess and Francisco Valero) has found a systematic 25 W m-2 discrepancy (between models and observations) for the amount of radiative energy reaching the surface. Some groups (Cess et al ., Ramanathan et al., Kiehl et al., and Pilewskie and Valero) have suggested that the discrepancy is due to unaccounted-for excess absorption in the atmosphere. He pointed out the very controversial nature of this issue, since many other groups (Stephens, Charlock et al., Arking, King, Ackerman, and others) do not find such a major discrepancy between observations and models.
CERES measurements may settle the issue of "excess" absorption by the atmosphere since it is the first satellite radiation budget experiment that will attempt to combine the TOA radiation budget with the surface radiation budget.
In addition, Ramanathan suggested the need for 3-D radiation modeling, as against plane-parallel modeling, as a way to understand the causes for the model-vs.-observation discrepancy.
Lastly, CERES on TRMM will provide a first look at the diabatic heating (latent plus radiative) in the tropical atmosphere.
Tuesday Afternoon, February 25, 1997
Robert Haskins (JPL) and Robert Atlas (GSFC) gave a joint presentation on "Prospects for Improved Weather Forecasts." Haskins opened with a discussion of improvements that may be achieved at the operational centers through improving input data, use of data, and forecast methodology.
He listed the space-based observational requirements and described the contributions to be made by EOS/AIRS toward improving weather forecasts. He pointed out that HIRS is an "undetermined system." He described the process of "ensemble" forecasting, saying that it works because the largest analyzed errors are observational and are not due to model errors. Ensemble forecasting has added a one-day improvement to the forecast process. He noted that "adaptive/targeted" observations are used to remedy forecast errors.
Atlas dealt particularly with the impact of NSCAT data on improving weather forecasts. He showed a sample of NSCAT-determined winds, which had the effect of correcting errors in locations and wind speeds of cyclones as against analyses conducted without the benefit of NSCAT winds. In the limited sample that has been studied, the improvements due to NSCAT were less significant in the northern hemisphere. Forecast centers that have used ERS-1 winds have shown a small positive impact on forecasts. Atlas said that he sees potential for improvements if wind profiles over the oceans can be obtained.
Eric Barron (Pennsylvania State University) and Soroosh Sorooshian (University of Arizona) presented "Assessing the Impacts of Climate on Regional Water Resources." Barron began by pointing out that human impacts are part of the goals of the USGCRP, and therefore the precipitation/hydrological cycles are important. The problem is how to go from global to regional scales that are more meaningful for their human impact.
He is concerned with embedding a mesoscale model in a GCM and has found that downscaling to the regional level does improve the forecast, both for a season and for a decade. In order for the nested-model approach to work, the GCM has to have good large-scale fields. He has found improvements in the precipitation fields but not the geopotential height fields or the zonal winds.
The improved precipitation analyses come from the improved topography and the improved physics in the regional modeling. He has also found that a neural net approach can work for the precipitation analysis. The technique shows promise for river-flow forecasts.
Sorooshian said that scale issues are important for the Colorado River basin. The mountains act as snow-water storage elements. He said that 98.7% of the precipitation in Arizona evaporates, whereas a far lesser amount evaporates over Louisiana. Unfortunately, the NEXRAD radar gives poor storm coverage in the southwestern U.S. because of the mountainous topography.
Sorooshian's group is looking at artificial neural network modeling, working with GOES data to provide a quarter-degree field of precipitation. Then, a stochastic approach can be used to distribute the precipitation at subgrid scale.
Richard Willson (Columbia University) presented "UARS/ACRIM II Results and the Long-term Solar Irradiance Database." He began by saying that sustained Total Solar Irradiance (TSI) changes have altered climate in the past. There is an "inverse" relation between solar activity and climate. Low solar activity has led to climate minima.
He said that an overlap strategy is needed for the satellite measurements in order to relate the TSI results over 100s of years.
He listed some of the relevant missions: UARS was launched in September 1991, and the ACRIM II data quality has been good. NPOESS will be launched in 2009 carrying an ACRIM instrument. SOHO/VIRGO data will be coming in March of this year.
Wilson said that there is always a problem in connecting results from various instruments on various spacecraft. The SMM/ACRIM-1 mission provided the longest period of spaceborne looking at the sun for TSI measurements. The upcoming problem will be the lengthy gap between the end of the EOS series 2 ACRIM II flight and the operation of ACRIM on NPOESS in 2009.
Steve Running (University of Montana) presented "New Applications of Remote Sensing for Wildland Fire Management." He started by saying that about $1 B is spent each year on wildfire suppression in the United States. Wildfire acreage is suddenly taking off. Dry trees don't decompose readily, and therefore about every ten years wildfires clean them up.
In one notable instance about $250 M was spent on halting the Yellowstone Park fires, but they didn't stop the fires--snow did!.
In current practice, fuels in the forests are mapped as a static parameter. Landsat, working with MISR, could give an up-to-date topography and fuels inventory. The instruments on the EOS PM-1 spacecraft will have a relevant "resistance" product. [Peter Mouginis-Mark pointed out that near-real-time fire and volcanic eruption data will both be provided by MODIS.] It was also pointed out that GOES 8 and 9 would provide data every half hour as against the two looks per day that might be had from MODIS.
Yoram Kaufman said that fire detection is not too important in the United States. It is really more important to monitor the growth of the fires. Reinhard Beer suggested that TES could be used to monitor flame temperatures if that turned to be useful.
Robert Harriss (Director, Science Division, Mission to Planet Earth, NASA Headquarters) presented "Planning the Next Generation U.S. Environmental Observing System." Harriss began by welcoming the new IDS investigators and said that he would be addressing issues of global change and what he called the "sustainability transition."
He stressed the great concern for the world population growth that may lead to 8 to 9 billion people in the next 50 years. The National Academy of Sciences Board on Sustainable Development, headed by Ed Frieman, is looking at the sustainability transition. The Board is to produce a "road map" for science and technology to face the problem. At the same time the national budget for the kind of multidisciplinary research that would address the population problem is shrinking.
There is now a National Environmental Monitoring Initiative with these elements:
[Interested people can check the world wide web at www. epa.gov/cludygxb.]
Harriss said that we need to move from global science to local/regional scales. Routine monitoring at the federal level costs about $600 M/yr, and yet these programs do not have high productivity--they are not highly policy relevant.
The Mid-Atlantic Policy Study inventories all activities related to environmental monitoring with the intention of fusing such activities to make them more efficient.
National Index Sites is a program to provide networks among such governmental programs as the NSF LTER and nongovernmental organizations (NGO) such as the Nature Conservancy. The intent is to get the science community to establish principles for operating integrated measuring sites.
The Environmental Report Card will tell the public the status of the environment. It was requested by Vice President Gore to be ready in 2001, and the first draft is due in 18 months. Harriss intends to spend a few thousand dollars with universities to have students prepare examples of the report cards that will lead to the final report.
The Next-Generation Monitoring Strategy for the United States will stress the contributions to be made by remote sensing. MTPE will lead the activity. There is a need to assess the activities that are now producing data that go unused. [For further information interested parties can check the world wide web or contact Don Pryor of OSTP: dpryor@ostp.eop.gov]
In the decade ahead there is to be an unusual convergence of issues including: slow productivity growth in the U.S. economy, major structural reform, and knowledge increasingly replacing land, labor, and capital. Also in the decade ahead all major policy issues will be confronted by new budget deficits. The good news is that the annual deficit is down; but the bad news is that the easy deficit reductions have been taken--increased taxes carry over from the Bush administration, and the DoD budget has been reduced probably as far as is likely or practicable.
Harriss asserted that through MTPE/USGCRP science there could be an increase in America's productivity by 1%/yr (or more) over the next ten years. He gave examples of how this increase could be achieved including: 10% from improvements in NCEP 14-day prediction skill; 30% from a shift to precision agriculture; improved energy demand forecasting; and growth of the commercial remote-sensing community, etc.
Wednesday Morning, February 26
Mike Mann (Deputy Associate Administrator for MTPE, NASA Headquarters) presented "MTPE/EOS Program and Project Updates." Mann started by reiterating the MTPE mission and goals, which are to develop understanding of the total Earth system (the mission) and to do so by expanding scientific knowledge of the Earth system; disseminating information about the Earth system; and enabling the productive use of MTPE science and technologies in the public and private sectors (the goals).
MTPE has the problem of translating diverse requirements into an integrated plan. It is driven by the USGCRP scientific requirements. There is an increasing stress on applications. Mann advised looking at the MTPE homepage to see the MTPE science research plan, the commercial strategy, the education strategy, and the program plan. This year's focus is on technology strategy.
Under the heading of "science planning" Mann reported a number of accomplishments and activities underway, including: Volume 1 of the MTPE Science Research Plan with 5 themes (published in September 1996); EOS Science Plan with 7 themes (now under- way--needs to be tied to the 5 MTPE themes); a USGCRP 10-year plan underway; and work in progress to integrate EOS science with other MTPE science.
Mann listed the five MTPE science themes:
Describing the MTPE program architecture, Mann said that the Earth System Science Pathfinder (ESSP) program is part of MTPE and will continue. It aims to have one low-cost/short-development-time spacecraft launch per year. Regarding in situ measurements, he said that there needs to be integration of platforms to include piloted aircraft and uncrewed airborne vehicles (UAV). UAVs are getting new emphasis.
He then discussed the international Earth observing programs. Japan is offering to step up its role. The international partners overall are putting about $4 B directly into the program, with a roughly equal amount in complementary activities, whereas NASA is spending about $7 B.
Changes in MTPE have brought out its increasing relevance and flexibility. The EOS program funding has gone from $17 B in 1990 to $7 B in 1997. We now recompete the IDS teams every three years, and there will be an Announcement of Opportunity (AO) to recompete the instruments for the second EOS series. There is now an aggressive move toward small satellites. The New Millennium Program (NMP) looks toward new technology. After the AM-1 launch the Delta-launched satellites will be the largest class of spacecraft in MTPE/EOS. We are heading toward formation flying. We will have an aggressive scientific research and applications program. New ways of doing business have us heading toward about 3.5- year mission-development times from the current 7-to-8-year pattern.
There has been a technology development transition. MTPE is now responsible for NMP, the Small Satellite Technology Initiative, and the Commercial Remote Sensing Program at the Stennis Space Center. Headquarters Code X has been eliminated, with its responsibility being transferred to the various NASA "enterprises." The "instrument incubator program" carries development through to laboratory or aircraft demonstration of feasibility. An advanced geostationary platform concept has been added to MTPE. As part of the Integrated Global Observing Strategy (IGOS) six pilot projects have been identified for consideration as international efforts.
MTPE has been reviewing recommendations that have come recently from the Earth System Science and Applications Advisory Committee (ESSAAC). There were three recommendations (necessarily paraphrased here):
Mann then showed the proposed MTPE response to the ESSAAC recommendations:
There are to be study teams functioning in three phases: Phase 1 has 7 study teams and was to report by the end of February. Phase 2 is to assimilate the results of the 7 teams of Phase 1. Phase 3 will involve a broadly-based external review process to support proposed program changes. IWG members are participants on each of the study teams.
Mann then moved to a review of other MTPE programs.
Thirteen proposals have been received in a second round to participate in ESSP. Selection is due by mid-March, and three awards are to be made in April. There is to be a first launch by March 2000, and one launch per year is planned thereafter. The third selection will be a backup, in case one of the first two selectees falters.
The New Millennium Program (NMP) will focus on land observations, using an advanced land imager (ALI). EO-1 is to be launched in May 1999, and EO-2 is to follow in the spring of 2002.
EOSDIS Federation planning is taking shape. The Earth Science Information Partners (ESIPs) will constitute a "working prototype" of the Federation. The existing DAACs are undergoing a certification process.
Turning to the MTPE budget, Mann said that there is to be a $50 M increase in FY 98. This, coupled with efficiencies implemented during the past year, permits the addition of SeaWinds II to the program. In the "big picture," MTPE is experiencing budget stability, although "earmarks" are having a significant impact. Congress is putting increasing emphasis on near-term applications. We need to step up our Science Outreach to demonstrate the usefulness of MTPE.
At the end of Mann's presentation there were a few comments. Dennis Hartmann said that "science" seems to have a decreasing percentage of the NASA budget. Bob Harriss said that we are losing "enormous" opportunities to be scientifically productive because the NASA grants program is going down; polar research, in particular, is dropping.
There was a question about the MTPE education program, and Ghassem Asrar said that the program has its own budget for the first time. It was suggested that it would be a mistake to cut back on the availability of science data to the public, and Mann replied that we need to know the cost of providing this access.
Mark Abbott (Oregon State University) and Ed Frieman (Scripps Institution of Oceanography) presented, jointly, "National Research Council/NAS: Recent Changes." Frieman, leading off, described the National Academy of Sciences (NAS) Board on Sustainable Development (BSD) and then gave the structure of the National Research Council (NRC). The Policy Division of the NRC includes: GUIRR, COSEPUP, STEP, and the BSD. The BSD was created by request of OSTP head, John Gibbons, as a body that would interface with OSTP and also interface with the President's Council on Sustainable Development. (PCSD).
The BSD is pursuing three overarching studies:
Following Frieman, Mark Abbott discussed changes at the NAS. There are now about 23 panels and committees at the NAS looking at matters related to USGCRP. There is an attempt to study interactions between NOAA and NASA that may lead to satisfying the CEOS requirements. Also being reviewed is the slowly evolving interplay between NASA and NOAA in the EOS series-2 development and how this will lead to support of NPOESS. Likewise, the interplay between NASA and NOAA in support of an advanced geostationary platform is under review. It is necessary to maintain harmony between the rapidly developing technology of NASA/EOS and the slowly evolving NOAA operational requirements.
William Chameides (Georgia Institute of Technology) presented "The Yangtze Delta of China: A Case Study of an Evolving Metro-Agro-Plex." (This is the work of a new EOS IDS study, the CHINA-MAP Project, that just received funding seven days before the presentation!) The study overall relates to aspects of world food production, NOx emissions, and SOx sources.
Chameides said that there is a strong correlation between industrial activity and food production. Air pollution is more than an urban problem. Pollution from urban areas is known to waft over food production areas. The presence of NOx and NOy is diagnostic of ozone production.
The biggest increase in pollution is expected to occur in East Asia, notably China. China is the world's most populous nation and is also the most rapidly developing. China's coal production will have doubled by the year 2010. They plan to move 500 million people from farms to urban areas! At this time China has adequate food production for its people, but they need to increase their grain supplies at the rate of 1%/year for the next 30 years. Unfortunately, pollution could have the effect of reducing their crop yields by the same amount.
Chameides listed some key issues that need to be addressed in his study: land-use change; SOx and NOx emissions as they lead to acid deposition; ground-level ozone; and climate change.
Rick Obenschain (Goddard Space Flight Center) gave the final formal presentation of the day. (The afternoon was set aside for viewing posters from the EOS instrument teams and the new IDS teams.) Obenschain's presentation was entitled "EOS Data and Information System Update."
Obenschain first described the functions of the administrative office, Earth Science Data and Information System, at Goddard, and described the "contents" of EOSDIS. He said that his presentation would focus on cost control, future directions for the EOSDIS Core System (ECS), long-term maintenance of cost containment, and Project-specific implementation. As part of a discussion of ECS "Replan and Implementation" he started with an overview. The transfer of TRMM responsibility from ECS to the DAACs has enabled reallocation of part of the Hughes staff to a focus on Landsat-7/AM-1/SAGE III. A stop-work order on TRMM had been issued on December 27 to Hughes, thus releasing 40 people. He noted that a major problem for ECS has been the high turnover rate for the people involved.
The Replan has two releases: B.0 for early mission instrument and algorithm calibration, and B.1 for full product generation and search and access capability. Development under the replan is proceeding apace to meet the mission schedules. A pre-Release B testbed will be available in mid-May, and a demonstration of critical Release B functions is scheduled for August 1997. A "points system" is now used to track the contractor's progress. About $5 M was transferred from ECS to provide TRMM operational capability.
Dave Glover (Woods Hole Oceanographic Institution) gave a brief response on behalf of the Data (EOSDIS) Panel, which had met just a few weeks earlier in Boulder, Colorado. Glover listed EOSDIS issues as: backup plans (they would be meeting with the SWAMP team on this); metadata (they plan a series of workshops on this); and the ESSAAC recommendations.
Glover said that the metadata issue is one source of tension between algorithm producers and database creators. There is disagreement on the number of metadata items required. The first workshop on this subject will try to establish B.0/B.1 metadata needs.
The Data Panel is concerned with the fundamental changes sought by the ESSAAC. They do not like the idea of assigning additional data responsibilities to the instrument PIs. The Panel has defined long-range goals for EOSDIS in the period following AM-1 and PM-1. Glover suggested that the Congress might not be happy with a data system that was available just to about 1000 NASA-related scientists.
There was general discussion after the short presentation by Glover. Hartmann asked whether there was still a desire for one-stop shopping, and if so, what is the cost? He urged that we still want data inter-operability.
Eric Barron asked about the sustainability of EOSDIS once the system is fully underway. What would the required staffing be for continuous operations? Obenschain replied that the operations staffing plans have been significantly reduced. Barron wondered whether the ESSAAC estimate of about 1600 people working at ECS was correct, and Obenschain said that he now sees the continuing need for just 100's of people for operations, including the DAACs and ECS.
Mike Freilich commented that the best understood system is the one we have now, and asked whether we can assign the costs appropriately now. Obenschain replied that individual requirements are supported by many elements of the system and that it is still not possible to trace individual requirements to cost. Mike Mann added that $1 B has already been spent to date, with $1 B to go, and that $600 M of this has already been committed. Thus the concern now is how best to spend the remaining $400 M.
Skip Reber stated that the Data Panel is trying to establish a first cut at EOSDIS as it will be after release B. He asked for feedback from the IWG on what is needed and noted that we still need to have the 250 standard data products. Harriss said that the ESSAAC thinks, wrongly, that we're inventing a new data system. The ESIPs have been given $12 M as an experiment to see how a system run by scientists might operate. This seems to be inadequate funding.
Obenschain said that to assure the functioning of the system, the B.0 release has to be ready by June 23, and a demonstration will be underway by the end of August. He also said that back-up plans are being considered right now. In answer to another question, Obenschain said that the data server is the "tall pole." "If it slips we lose our contingency."
Asrar urged that the PIs get involved directly in problem solving, rather than have a chain in which ESDIS works with ECS and then with the PIs. Freilich said that the problems should be stated directly, and then the scientists would respond.
Mark Abbott said that the system has to be organized so that when things break they have minimal damage effects. Timely delivery of products is the key to outside support. Obenschain's concern was that having a "B.0`" system with only limited output could hurt the program.
Jeff Dozier pointed out the underlying issue, that today's concerns are not new but, that what is new, is that now we have a crisis. We need to think about getting out of "management by crisis." He asked what we are doing to address the crisis concerns? He thought we ought to consider the cost of setting too high a reliability goal.
Wielicki questioned whether we are taking too big a step in trying to go beyond the V0 system. He asked whether it was true that V0 meets our requirements right now.
Dozier said that what we have now is a geographically dispersed system, but it is not a "distributed" system. The PIs can take care of their own parts of the system with the certainty that there will be "screwups," but also with the certainty that not everything will go wrong.
Barkstrom asked whether there are any concepts or criteria as to when to go to a true distributed system. Obenschain replied that somewhere in the period, August to September, there should be a decision. Mike Mann said a decision should be made earlier.
This session ended with Mark Abbott saying that in all this discussion there is an implicit issue: that there is always just one path into the data system. In fact, there could be multiple paths to get to the data--open back doors, shadow systems, ...
Thursday, February 27, 1997
John Hrastar (Deputy Director, MTPE Program Office, Goddard Space Flight Center) substituted for Robert Price, presenting "EOS Chemistry-1 Internal Study Results." He noted that the CHEM-1 mission costs are projected to run to something like $700 M. There had been a CHEM-1 study in February-to-May of 1996 to see if there could be a 50% reduction in mission costs. As part of this study several alternative mission configurations were reviewed. Then in May-to-June of 1996 an Implementation Assessment found that there should not be any changes in the instruments, but there were reasonable options for splitting up the instruments among spacecraft.
The Project has been exploring the possibility of developing a $40 M spacecraft to be the CHEM platform. A way of doing this has been to set up cooperative agreements with industry on a 50% cost-sharing basis. Final results of this process are due in November. (This same approach is also to be used to explore platform possibilities for the laser altimetry mission.)
So far, eight contractors have participated in the cooperative study. They have offered possibilities of having new designs, using existing communications buses, and using existing Earth-imaging buses. The conclusion of this effort is that industry can provide the necessary class of medium satellites. The biggest cost drivers have been shown to include such elements as single fault tolerance (for a five-year lifetime) and labor costs. As the spacecraft get smaller the instrument costs tend to dominate the mission costs. A positive result of the study has been the establishment of a very good technology/cost database.
The current baseline is the common spacecraft to be supplied by TRW, but a mixed fleet is still a good option. The TES and MLS PIs have been asked to do a quick study of adopting the "PI Mode" for their own instruments.
At this time a December 2002 launch of CHEM-1 on the common spacecraft remains the baseline. It appears to be the best choice, based on the economics. There is a problem, however, with the cost of launch vehicles. Still, a decision on this is not needed until next year.
Richard Holdaway of the UK warned that the Europeans are worried about the US uncertainties of how missions are to be carried out. Hrastar replied that we are on track for the common spacecraft. Mike Mann said that we'll be "locked in" on our decision by this summer.
Daniel Jacob (Harvard University) presented "Latest Progress and Future Plans for Tropospheric Chemistry." He said that he was representing a new tropospheric chemistry interdisciplinary science investigation (IDS), which addresses the heart of environmental science and policy. Tropospheric chemistry is very important in terms of its consequences for climate, but also very uncertain. Regional differences are very important because of the short lifetimes of airborne pollutants and their inhomogeneous sources. It is important to resolve their coupling with weather phenomena on synoptic scales, and there are very complicated feedbacks that must be considered.
In the new IDS study the plan is to put all relevant chemistry into a GCM, but focusing on O3 and SO4. Jacob described the chemical reactions leading to the formation of OH, the "cleansing agent" of the atmosphere, and added that the cycling of NOx is essential to maintaining the OH concentration. It is understood that transport of NOx from the stratosphere is of little importance. So far, NO measurements have come from aircraft flights, which offer very limited coverage. TES from EOS will be the preferred source of tropospheric ozone measurements and the precursor gases.
Fossil fuel combustion plus biomass burning are the sources of about 70% of NOx emissions. The measurements of HNO3 and "PAN" are still poor. There is a question as to whether acetone is the source of HOx in the upper troposphere.
Anthropogenic processes are the major source of sulfate aerosol formation. The conversion of SO2 to H2SO4 takes place primarily in clouds. There is a big microphysical issue as to whether the conversion leads to the formation of new cloud particles or whether there is just condensation on existing particles.
Harvard's Chemical Transport Model (CTM) includes transport from the GISS GCM. Currently, the model lacks feedback from chemistry to meteorology, but this IDS investigation will be looking at feedbacks between chemistry and climate. Operational versions of the model are expected to be available by May, and then it will be possible to evaluate model results against observations. Heterogeneous chemistry in clouds will be modeled. There is a major problem in accounting for rainfall scavenging of aerosols.
Jack Fishman (Langley Research Center) presented "Progress on Measuring Tropospheric Ozone from TOMS/SAGE and Other Satellite Data." He said that tropospheric ozone is increasing worldwide, but not uniformly. The amount of increase depends on season and locale. He has been able to identify enhancements, which he can attribute to biomass burning and transport, in the total ozone measurements from TOMS for 1985 and 1986. He has also found tropical ozone enhancements over the mid-Atlantic, which he has been able to relate to biomass burning in Africa. He noted that some work by Anne Thompson (GSFC) has shown that clouds affect the ozone retrievals so that actual enhancements may be less than previously calculated by about five Dobson units. An interesting study result was the ability to follow a mass of ozone-polluted air as it moved from the northeast around a Bermuda high over the ocean and then back to the southeast.
Fishman believes that a geostationary platform would be ideal to capture the heterogeneity in time and space of the world's tropospheric ozone. He envisions an instrument that would measure carbon monoxide, nitrogen dioxide, and sulfur dioxide, as well as tropospheric ozone, all from a communications satellite on geostationary orbit. The measurements would have 0.5-to-2.5-km resolution and could be made every 15-to-20 minutes.
Brian Toon (Ames Research Center) presented "Aerosol and Climate Interactions." (He was reporting on a new EOS IDS project.) He gave three reasons as to "why aerosols matter":
Toon emphasized that dust makes up a type of aerosol that is more important than sulfate aerosol for climate effects. Dust optical properties are widely varying, and they can have regional effects that are as great as those due to cloud forcing in the infrared. Among direct effects of aerosols on climate, uncertainties arise due to: (i) imprecise knowledge of aerosol optical depths, especially over land; (ii) changes in aerosol properties as they move away from their source region; and (iii) insufficient comparisons between model results and observations.
Toon discussed the "Twomey effect" (named after Sean Twomey). The effect relates to the phenomenon wherein, as aerosol concentrations increase, cloud droplet concentrations increase and droplet size decreases, leading to increases in cloud albedo. This is of interest because aerosol particles can serve as cloud condensation nuclei (CCN), thus leading to changes in cloud droplet concentrations. Other aerosol-related phenomena cited by Toon were these: (i) stratus cloud heights change over time as CCN are depleted; (ii) cirrus cloud characteristics show only a slight dependence on the presence of aerosol particles; (iii) in the presence of soot particles, cirrus clouds increase in area and optical depth; and (iv) clouds are sensitive to the abundance of CCN at low concentrations of CCN.
Toon reported successful modeling of the El Chichón and Mt. Pinatubo volcanic eruptions. The model results show, for example, the observed spread of the Mt. Pinatubo aerosol cloud into the southern hemisphere. Attempts to model dust events over the Persian Gulf have shown the need for high-resolution calculations.
Jim Hansen (Goddard Institute for Space Studies) presented "Forcing and Chaos in Interannual to Decadal Climate Change." (As an aside he noted that students in the GISS "outreach" program have been involved in testing the climate model by comparing it with climatological data and data for the study period, 1979 to 1996.)
His principal conclusions are that there is a clear indication in the data of a climate response to both natural and anthropogenic forcing, and that unforced atmospheric variability (chaos) is the principal source of change. Global annual mean temperature is strongly driven by radiative forcings, but on the average, about 3/4 of local variability is chaotic on the 17-year period of investigation.
In his modeling the oceans are represented by time-varying sea-surface temperatures (SSTs) and fixed sea ice. In assessing his atmospheric model, he found that the lower stratosphere was not well represented, and the forcing of the oceans by the atmosphere is deficient. In intercomparisons, he found that his model is "more or less" state of the art. Climate forcing from tropospheric aerosols was not included in the model.
Hansen credited the SAGE instrument with giving invaluable ozone profile information in the crucial tropopause region, but stated that verification is still needed, as well as data at lower levels.
In computing net forcing he took into account stratospheric aerosols, ozone, greenhouse gases, and solar irradiance. He has found that the net forcing in the study period was less than 1/2 W m-2. Vegetation changes were not included in the 1979-96 study period, but a sensitivity study using preindustrial vegetation showed that anthropogenic land use causes a global forcing of about -0.4 W m-2. Significant impacts on temperature occur in the regions of vegetation change.
In listing findings, he said that ozone changes and stratospheric aerosols have large demonstrable impacts on atmospheric temperature. Also, he has found evidence for a disequilibrium in the planetary radiation balance, presumably due to greenhouse gases added to the atmosphere prior to the study period.
Norman Miller (Lawrence Livermore Laboratory) presented "Coupling Global Climate Models to Regional Hydrometeorological Models." (Miller is another one of the new IDS PIs, collaborating with Jenwin Kim.) His work is greatly concerned with establishing soil moisture feedbacks. He has a "homogeneity algorithm" that converts fine-resolution imagery to coarse resolution. He uses multiple process models to determine impacts and make assessments of the effects of climate changes. As part of his regional work he has been able to match simulated precipitation with observed precipitation for northern and central California. He has found that his models do a good job on soil moisture simulations. Looking at the Russian River basin, he has found that his models are successful in simulating the river flow.
Some of his work is related to Eastern Asia hydroclimatologic research. There the goal is to understand the impacts of global climate variability on the hydrological climate, on the ecosystems, and on agriculture. For this work he uses 60-km topography, but then does "model nesting" down to 20 km, working with six-hour updates.
Cynthia Rosenzweig (Goddard Institute for Space Studies) presented "Assessing the Impacts of Climate on Regional and Global Agriculture." Hers is another of the new EOS IDS projects. Her study deals with the impacts of climate on agriculture and related areas such as ecosystems and fisheries. Her study objectives include providing a coordinated framework to assess and predict climate impacts of large-scale fluctuations in important food-producing systems around the world; improving forecasts of climate changes; and investigating means of mitigating the negative impacts of climate variability.
She started her talk by calling attention to improvements that are soon to be coming in predictions of ENSO events.
Two important concepts to keep in mind are these: 1) There is a "user pull" for information on climate impacts from farmers and related interests such as the industries that deal with fertilizers, storage, and processing, and also the consumers; 2) There is a need for improved communications on data and research activities among scientists who come from many different fields.
Rosenzweig's project has links to NOAA and to the International Research Institute for Climate Prediction. Her team makes use of Cane's work on ENSO forecasting, Rind's GCM studies, and Tucker's remote sensing of vegetation change. Study tasks include: test predictions vs. historical studies; develop near-real-time prediction tools; and test mitigation strategies.
Study regions for the project include the U.S. cornbelt, Northeast Brazil, Mercosur, and Zimbabwe. (There is also interest in Southern Brazil, Uruguay, and Argentina.) These regions are chosen for their differing vulnerabilities to climate change.
Scales of analysis, both spatial and temporal, have to be taken into account. Usefulness of information can be dependent in different ways on spatial scales that range from pixel size to GCM scales. Daily and weekly temporal scales affect crops in different ways. Rainfall timing is especially important to crop development.
The project has conducted experiments with GCM ensemble runs. There have been experiments using global observed SSTs, others using tropical Pacific observed SSTs, and others using Cane-Zebiak-predicted SSTs.
Tucker's work has identified a correlation between NDVI and SST anomalies (looking at the period 1982 to 1990). A strong correlation has been found between El Niño temperatures and rainfall effects on corn yield in Zimbabwe.
Anthony Michaels (University of Southern California) presented "Climate Variability and Insured Risk: What is the Value of Remote Sensing to the Global Insurance/Reinsurance Industry." Michaels focused on what he called the Risk Prediction Initiative (RPI), saying that the impetus for the Initiative was the big losses incurred by insurers in 1992, the year of Hurricane Andrew. His work is in the area of trying to bring scientific knowledge into the risk calculations of the insurers.
Michaels gave some interesting insights about the functioning of the catastrophic insurance organizations. In their risk considerations, the catastrophic insurers assume that there will be only two major events per year. A loss of $100 B is regarded as the canonical loss. The total capital of the entire insurance industry is only $250 B.
Jim Yoder (NASA Headquarters) presented "Early Results from the OCTS Mission." OCTS is the Ocean Color and Temperature Scanner and is now operating on board the ADEOS spacecraft along with POLDER.
Yoder gave a brief listing of related missions and instruments: NSCAT (on ADEOS) is providing ocean surface winds; AVHRR Pathfinder is providing SSTs; POLDER provides aerosols plus global ocean chlorophyll (started in November 1996); and SeaWiFS (projected to be launched in 1997) will provide global high-resolution chlorophyll measurements.
Yoder explained that the usefulness of ocean color lies in the fact that chlorophyll reflects green light, and thus its signal can be used to estimate plant life in the ocean, ocean productivity, and nutrient uptake. The key remote-sensing problem is that atmospheric corrections are needed since 90% of the signal reaching the satellite comes from the atmosphere.
Yoder listed some of the characteristics of POLDER and OCTS. POLDER has 8-km spatial resolution whereas OCTS has ~1-km resolution. He also noted that there has been a 10-year data gap from the last satellite-based ocean color measurements until now.
POLDER and OCTS are now still in their calibration/validation (cal/val) stage. The SeaWiFS and MODIS science teams are assisting with cal/val for both POLDER and OCTS. They use buoys for this support activity. Global products from the two instruments should be available by the summer or fall of this year.
Yoder wryly pointed out that SeaWiFS activities have now gone on for ten years with the first flight yet to come. The SIMBIOS project has been formed by NASA to provide for the smooth merging of data from the sensors previously mentioned plus future sensors that may come along. Also, a new coordinating committee has been set up to provide an agreed Internet program for ocean color with the intent of reducing redundancy.
Ghassem Asrar (NASA Headquarters) presented "Direct Broadcast of EOS Data." He reviewed some of the technical aspects of direct broadcast (DB), saying, however, that DB is lagging behind other aspects of the EOS program. On AM-1 only MODIS will have the DB capability, whereas on PM-1 all the instruments will have the capability. Landsat's capability will be more like Direct Downlink, in that it will require pointing.
Peter Mouginis-Mark, University of Hawaii, has adopted a DB receiving system that should be widely usable because of its relatively low cost. A 5-m programmable tracking antenna is used along with special software to process the downlinked data. The U of H antenna costs about $650 K, including the dish, the pedestal, and ingest capability. It is estimated that about 100 users of the system will be in place around the world. There will be a planning meeting on April 14-17 at the University of Hawaii.
A comment from the audience was that the system can handle 150 Mb/s. Also, to save money it would be possible to go to a 3-m dish that would still be adequate to receive the MODIS signal, and then the system cost would be about $450 K. Operating costs would be about $100 K/yr.
With this final presentation the meeting was adjourned.